Toward Data‐Driven Applications in Lithium‐Ion Battery Cell Manufacturing

Abstract

Production chain of lithium‐ion battery cells is a highly complicated system with manifold process–product interdependencies and high sensitivity to ambient conditions. This complexity makes it harder to control and regulate economic and environmental target criteria (e.g., product quality, cost, and energy demand). Therefore, it is necessary to develop a holistic system understanding and to identify and evaluate the interactions between the process steps within the production chain of battery cells and their effects on the relevant cell properties. In this study, a data‐driven concept to acquire relevant data along the production line, including technical building services and cell diagnostics, is presented. Furthermore, it describes the combination of automated and manual data acquisition as well as merging of data from different sources, of different communicational protocols, and of different formats toward accessibility, convenient data management, and visualization.

Artificial Intelligence Investigation of NMC Cathode Manufacturing Parameters Interdependencies

“…Some prospective papers discuss the potential of ML methods to help boosting the discovery of efficient solid electrolyte interfaces. [30,31] The latter focuses on the macroscale. [30,31] The latter focuses on the macroscale.…”

“…[29] Few recent works reported automatic ways to collect, store and analyse experimental data in order to allow a better control of the battery production chain thanks to ML methods such as decision tree or random forest. [30,31] The latter focuses on the macroscale. Therefore, the development and application of ML tools able to predict the correlations between the manufacturing parameters and the final properties of LIB electrodes remains elusive.…”

“…Some prospective papers discuss the potential of ML methods to help boosting the discovery of efficient solid electrolyte interfaces . Few recent works reported automatic ways to collect, store and analyse experimental data in order to allow a better control of the battery production chain thanks to ML methods such as decision tree or random forest . The latter focuses on the macroscale.…”

Accelerating Battery Manufacturing Optimization by Combining Experiments, In Silico Electrodes Generation and Machine Learning

“…[1][2][3][4][5] However, their performance, durability, recyclability and CO2 fingerprint require further improvement to make feasible this transition. This can be achieved thanks to materials design, [6][7][8] innovative manufacturing [9][10][11] and manufacturing optimization, [12][13][14] or most likely, thanks to a combination of all of them. Regarding the latter, both experimental and modelling approaches can be used to carry out its optimization.…”

“…The blooming Artificial Intelligence (AI) field promises to accelerate the manufacturing optimization by revealing patterns hardly recognizable by "classical" analysis methods. 14,[32][33][34][35][36][37] As they do not rely on physical models, the feasibility of this approach depends on the capability to generate high quality datasets (from experiments, physical models or both of them simultaneously) complete enough to describe the battery manufacturing, which most likely represents the limiting step to develop AI models. Takagishi et al recently reported a machine learning (ML) approach in which the datasets were built by randomly generating in silico electrode mesostructures composed of only AM particles coupled with a zerodimensional electrochemical model to calculate the charge/discharge specific resistance.…”